Dear R-SIG-GEO We would like to inform you that version 1.5.0 of `sits` package in now on CRAN.
`sits` is an end-to-end, TRL 9, operationally-tested package for big Earth observation data analytics using satellite image time series and machine learning. Noteworthy upgrades in version 1.5.0 include: (a) Support for Sentinel-1 and Sentinel-2 collections in Copernicus Data Space Ecosystem; (b) Support for Sentinel-1-GRD and Sentinel-1-RTC collections in Microsoft Planetary Computer; (c) Support for Digital Earth Africa products SENTINEL-1-RTC, LS5-SR, LS7-SR, LS9-SR, ALOS-PALSAR-MOSAIC, NDVI ANOMALY, DAILY CHIRPS, MONTHLY CHIRPS and DEM-30; (d) Improved performance on GPU-based classification of deep learning models; (e) Merging Sentinel-1 and Sentinel-2 data cubes; (f) Include DTW distance when building self-organized maps (SOM) for training data quality control; (g) New `sits_reduce()` function for multi-temporal statistics; (h) New functions `sits_sampling_design()` and `sits_stratified_sampling()` to implement best practices for classification assessment based on the best practices proposed by Olofsson et al. (2014). Documentation is available as an on-line book, available at https://e-sensing.github.io/sitsbook/. `sits` relies on the strong community spirit of the R community and in particular the r-spatial team. We acknowledge our debt to Edzer Pebesma (`sf/stars`), Marius Appel (`gdalcubes`), Robert Hijmans (`terra`), Tim Appelhans (`leafem`), Jakub Nowosad (`supercells`), and Martijn Tennekes (`tmap`). We are grateful for the work of Dirk Eddelbuettel (`Rcpp/RcppArmadillo`) and Ron Wehrens (`kohonen`). We are much indebted to Hadley Wickham for the tidyverse, Daniel Falbel for the `torch` and `luz` packages, and the RStudio team for package `leaflet`. The CRAN team (Uwe Ligges and prof Ripley) has been very supportive. We also thank Roger Bivand for his benign influence. Without the commitment of the r-spatial community, development of `sits` would have been impossible. It is amazing and impressive how the packages of the r-spatial community fit seamlessly like LEGO blocks. Arguably, this is due to amazing consistency and capabilities of the `sf` package. It has become a sound basis for most other r-spatial packages, including `sits`. Kudos to Edzer and all contributors! Best regards Gilberto ============================ Prof Dr Gilberto Camara Senior Researcher National Institute for Space Research (INPE), Brazil https://gilbertocamara.org/ ============================= _______________________________________________ R-sig-Geo mailing list R-sig-Geo@r-project.org https://stat.ethz.ch/mailman/listinfo/r-sig-geo
